1. Computers
  2. Display Drivers
  3. Graphics Cards
  4. Memory
  5. Motherboards
  6. Processors
  7. Software
  8. Storage
  9. Operating Systems


Facebook RSS Twitter Twitter Google Plus


Phoronix Test Suite

OpenBenchmarking.org

Zack Rusin Talks About Gallium3D's TGSI IR

Mesa

Published on 28 October 2010 06:41 PM EDT
Written by Michael Larabel in Mesa
2 Comments

Lately there's been a lot of talk about Gallium3D's IR known as TGSI, or Tokenized Gallium Shader Instructions, and attempts by some to replace this intermediate representation. Efforts toward improving TGSI are not particularly new, but it's been going on for a while and then just earlier this month a new shader and compiler stack was proposed by LunarG. As part of the LunarGLASS proposal, the LLVM IR would be used as a replacement to TGSI.

Zack Rusin of VMware (formerly with Tungsten Graphics) has just written a blog post about Gallium3D's Tokenized Gallium Shader Instructions for those interested in more information regarding the technical side of the IR and its current shortcomings and hurdles being experienced by these open-source GPU driver developers.

Zack's blog post about the IR can be read on his Blogger page. As some takeaway comments for those not interested in all of the technical details but just an overview of the situation, "[Tungsten Graphics'] lofty goal of making the IR so much easier and quicker to adopt took a pretty severe beating. It's especially disappointing since if you look at some of the documentation e.g. for AMD Intermediate Language you'll notice that this stuff is essentially Direct3D assembly which is essentially TGSI (and most of the parts that are in AMD IL and not in TGSI are parts that will be added to TGSI) . So they have this code. In the case of AMD it's even sadder because the crucial code that we need for OpenCL right now is OpenCL C -> TGSI LLVM backend which AMD already does for their IL. Some poor schmuck will have to sit down and write more/less the same code."

The most important paragraph in Zack's blog post is perhaps, "So we're left with Free Software developers who don't have access to the Direct3D functional spec and who are being confused by the IR which is unlike anything they've seen (pre-declared registers, typeless...) which on top of it is not easily transformable. TGSI is very readable, simple and pretty easy to debug though so it's not all negative. It's also great if you never have to optimize or transform its structure which unfortunately is rather rare. If we abandon the hope of having the code from Windows drivers injected in the GNU/Linux drivers it becomes pretty clear that we could do better than TGSI. Personally I just abhor the idea of rolling out our own IR. IR in the true sense of that word. Crazy as it may sound I'd really like my compiler stuff to be written by compiler experts. It's the main reason why I really like the idea of using LLVM IR as our IR."

About The Author
Michael Larabel is the principal author of Phoronix.com and founded the web-site in 2004 with a focus on enriching the Linux hardware experience and being the largest web-site devoted to Linux hardware reviews, particularly for products relevant to Linux gamers and enthusiasts but also commonly reviewing servers/workstations and embedded Linux devices. Michael has written more than 10,000 articles covering the state of Linux hardware support, Linux performance, graphics hardware drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated testing software. He can be followed via and or contacted via .
Latest Linux Hardware Reviews
  1. CompuLab Intense-PC2: An Excellent, Fanless, Mini PC Powered By Intel's i7 Haswell
  2. From The Atom 330 To Haswell ULT: Intel Linux Performance Benchmarks
  3. AMD Radeon R9 285 Tonga Performance On Linux
  4. Apotop Wi-Copy
Latest Linux Articles
  1. AMD Moves Forward With Unified Linux Driver Strategy, New Kernel Driver
  2. MSI: Update Your BIOS From The Linux Desktop
  3. NVIDIA vs. AMD 2D Linux Drivers: Catalyst Is Getting Quite Good At 2D
  4. 15-Way GPU Comparison With Mesa 10.3 + Linux 3.17
Latest Linux News
  1. NVIDIA's NVPTX Support For GCC Is Close To Being Merged
  2. KDE's KWin On Wayland Begins Using Libinput
  3. Khronos Releases OpenVX 1.0 Specification
  4. Linux Kernel Working Towards GNU11/C11 Compatibility
  5. Ubuntu 15.04 Is Codenamed After A Monkey: Vivid Vervet
  6. Following GCC, Clang Looks To Default To C11
  7. Users/Developers Threatening Fork Of Debian GNU/Linux
  8. Linux 3.18-rc1 Released One Week Early With Many Changes
  9. The VC4 Gallium3D Driver Is Still Moving Along For The Raspberry Pi
  10. Direct3D 9 Support Might Land Within Mainline Mesa 3D Drivers
Latest Forum Discussions
  1. Bye bye BSD, Hello Linux: A Sys Admin's Story
  2. Users/Developers Threatening Fork Of Debian GNU/Linux
  3. HOPE: The Ease Of Python With The Speed Of C++
  4. NVIDIA Presents Its Driver Plans To Support Mir/Wayland & KMS On Linux
  5. AMD Is Restructuring Again, Losing 7% Of Employees
  6. Open-Source AMD Fusion E-350 Support Takes A Dive
  7. Upgrade to Kaveri, very slow VDPAU performance
  8. ChromeOS Drops Support For EXT2/EXT3/EXT4 File-Systems